Bob Hartley and team shooting percentages

So, if you’re a Flames fan, you’ve heard a lot about shooting percentages over this past season. That’s to be expected: Calgary did, after all, experience an unprecedented amount of success… at the exact same time they saw a great rise in their shooting percentage, all the way up to second in the NHL.

It was last season that saw Bob Hartley win the Jack Adams Award, too. He had a team many predicted would be among the worst in the league, and instead, they made the playoffs. It’s the exact same cause for Patrick Roy to have won the trophy the year before – even though, predictably, the Avs fell off a cliff the following year.

Is the same going to happen to Hartley’s Flames? Or is it possible that what Hartley accomplished last season is repeatable?

It’s all very random

As we saw this past season, there really isn’t a correlation between how good your team is at shooting, and how good it is at winning. The Tampa Bay Lightning were the NHL’s top-shooting team, and they finished fifth in the league; the Flames, meanwhile, were second in line shooting-wise, but finished 16th. 

The Chicago Blackhawks went on to win the Stanley Cup, but in the regular season, they were the fifth-worst shooting team in the NHL, and still finished seventh in the overall standings. The New York Rangers had a team shooting percentage of 9.60%, and the Columbus Blue Jackets, 9.59%: one team won the President’s Trophy, and the other never even so much as sniffed the playoffs.

There are a lot of different reasons for this, because hockey is a fairly complex game. On-ice personnel plays a large factor; in particular, goaltending. Some players are just naturally better shooters than others: guys like Alex Tanguay, Jiri Hudler, and maybe even Sean Monahan fall under this category. Luck plays its role, too: in a game as fast-paced as hockey, sometimes weird years just happen.

The same goes throughout all 10 full seasons of Bob Hartley’s NHL coaching career.

hartleycareerranks

Hartley’s top team (and, yes, the Stanley Cup Champion Colorado Avalanche) was a decent shooting team compared to the rest of the NHL, but his best relative shooters – this year’s Flames – didn’t perform that well in the standings. Hell, even one of his worst shooting teams had a pretty good season, finishing fourth in the league.

There’s more correlation when looking at just plain old shooting percentage and comparing it to a team’s place in the standings.

hartleycareersh

Hartley’s worst shooting team was his worst performing team, while his best was, well, the best. Everything in between isn’t quite as linear, and there are a few outliers, but in general: the more accurate at shooting they were, the better his teams tended to do. 

That worst shooting team was the 2013-14 Flames: a 9.19% shooting percentage, and a 27th place finish. One year later, with largely the same personnel, the Flames became a 10.52% shooting team – his fourth-best shooting team ever – and finished 16th.

How did such a poor shooting team become a great one? The real changes to the lineup were Brandon Bollig, Johnny Gaudreau, Josh Jooris, a slightly more experienced Sean Monahan, Mason Raymond, and no more Mike Cammalleri, Brian McGrattan, or Lee Stempniak. Is that enough to affect that great a change?

Brent Sutter vs. Bob Hartley

Brent Sutter coached the Flames for three seasons, and Hartley took over for the next three. The two shared a fair number of players, in particular between the 2011-12 and 2012-13 years.

sutterhartleysh

Hartley has had, overall, the more accurate teams – and the worse ones. Sutter was able to achieve roughly what Hartley did with his editions of the Flames, but he didn’t have to rely on them shooting as accurately in order to do it. Hell, if the Flames, under Sutter, had had as much accuracy and luck as they did under Hartley’s most recent year, the playoff drought may have been broken earlier (and the rebuild may have been delayed even further).

Overall, though, the Flames haven’t been fantastic shooters in recent seasons – just the last one. When they did about as well as Sutter’s teams did.

When the Toronto Maple Leafs made the playoffs in 2012-13, they were the top shooting team in the league, at 11.47%.

The next season, they fell to a less nice 9.69% – and were 23rd in the NHL, under the same coach.

History doesn’t dictate they repeat, and it also kind of dictates they need to. This is where we’ll see just how big an impact the offseason acquisitions really have: if they can help keep this team afloat when it’s not shooting at 10.52% again. 

Via QuantHockey – where these numbers are coming from – the average across the league tends to be at about 8.90%. The Flames have been mostly above average, but it’s only really helped them once.

An overall improvement would be ideal… but hey, if they can keep this part up, too, nobody’s going to complain. It just shouldn’t be a factor relied upon, because over the long-term, it doesn’t exactly work.

  • Christian Roatis

    SH% is probably one of the only few truly random metrics. There is literally no way to predict it. Every scoring shot is millimetres away from not going in, and vice versa.

    Good piece, Ari.

  • JBudd

    I’d be interested in seeing this but for team save %. My gut tells me a coach’s system has more ability to lower other teams shooting percentages by keeping them to the outside than to increase your team’s ability to score. I have a feeling the numbers may not back up my gut though.

    • DestroDertell

      Copy-paste of what I posted earlier this week on MS&G:

      “flames defense allowing [most] shots from low percentage areas in a myth. They were 25th last year at allowing high-danger chances and 27th at allowing scoring chances. Mostly because of Russell and Wideman being bottom pair defensemen dealing with top 4 competitions and Engelland playing more than 3 minutes every night.”

  • JBudd

    I’m going out on a limb, but there are a couple of factors that helped the Flames last year. The Flames have some very good passers. They tend to shoot on net less, but could have a deceptvely good shot. Johnny and Hudler are examples.

    They also took fewer shots per game than 27 other teams. That could be a product of Calgary being more choosey in the shot taken, resulting in a more dangerous shot.

    A weird byproduct of playing from behind is the 6-on-5 with more than a minute remaining. You are going to tend to shoot less, because the risk of losing possession likely results in a goal.

    Just some theories.

    • KACaribou

      Quote from Jonathon Quick’s article on Snipers 101:

      “There’s obviously a ton of emphasis on puck possession in the media and in NHL locker rooms with the advanced stats movement really growing.

      But I think it’s about the kind of possession you have. Some teams might have a lot of puck possession in your zone, but they’re really not in threatening positions. They might be cycling the puck around on the perimeter and throwing some stuff on net, but that’s pretty easy to deal with.”

      My eye tells me that a lot of Flames goals last year were scored from high percentage areas (the threatening areas of the ice)and therefore resulted in a high scoring percentage.

  • ngthagg

    I’m hoping one of the more experienced stats guys can answer a question for me: why, in the age of Corsi, don’t we calculate SH% with missed and blocked shots included?

    The increased sample size should help eliminate some noise, giving us a number closer to a player’s true SH%. And that, in turn, will be a more accurate number to make predictions from.

    Has anyone done this? If so, where can I read about it? If not, is anyone up for the challenge?

  • DestroDertell

    At even strength, even with the 2nd highest SH% in the league, the flames had a negative goal differential and were only ranked 15th at scoring goals. They flames were 19th at allowing goals in the same situations in spite of Hiller’s very good adjusted sv%.

    Good thing they were the least penalized team in the league. If having bad possession numbers to control accuracy on both side of the ice was the plan, it was not a good plan.

      • KACaribou

        Being good at drawing penalties sounds like the Flames are a team of divers.

        They CAUSED a lot of penalties because they were on the puck and often in scoring position.

        Winning in extra time was also no fluke. They could put 4 players out there who were fast, controlled the puck and could put it on net. Fast and skilled is becoming more important than big and truculent.

          • KACaribou

            Good point, though I wasn’t really thinking of playoffs – just getting there. I am mostly tired of some stats which seem to belittle the Flames’ effort this past season like it was some kind of fluke. I watched those games, every one of those games. What a great team. What a great coach! Can’t wait to see how much better Johnny Hockey, Moneyhands, Superstar Bennett, and Train 79 are this coming season.

    • The GREAT Walter White

      “Goal differential” is by far my favourite statistic!!!

      What I like about it the most is that the smaller the sample size the better it becomes at predicting the outcome of games. In fact; using this stat for any single game has an almost 100% accuracy in predicting the outcome……what other “advanced” stat can do that?

      Exactly…

      WW

      • DestroDertell

        Here’s a suggestion: look up what the word predicting means. Outscoring is how you win game. But it doesn’t mean you outscored a team once you’ll do it again the next time you play them again. Thus it doesn’t have 100% accuracy in predicting the outcome.

    • Section205

      5on5, Flames were disciplined defensively, but not agressive in taking away possession. They were collapsing, blocking shots and being the least penalized team. (opposite of Winnipeg)

      I believe a trade-off of this approach is you give up more even strength shots against (7% of these go in) but far fewer shorthanded shots against (11% of these go in). As a result they were out shot 5on5 and thus likely outscored 5on5, but they gave up fewest SH goals in the league.

      Drawing penalties, they were 10th. Many teams had more power plays. Flames PP was pretty good in the opportunities they had.

      As least penalized team, Flames had massive shot advantage on special teams, where shooting percentages are highest.

  • Derzie

    So the opening statement is how SH% is not related to winning. I stopped reading right there. Why would one care about a stat if it doesn’t indicate something important?

  • KACaribou

    Some credit has to be given to the Flames goalie coach that does a good amount of research on the opposing goalie and educates the team on it as part of pre-game preparation.

    Hartley did say at the beginning of the season that scoring might be an issue, so he and the coaching staff came up with something that utilizes their defense and it worked well.

    My point being, a lot of this was by design and it wasn’t completely a fluke as some analysts would paint it. Now, the degree in which their SH% ranks at may fall (there is some luck involved after all) but I predict they’ll still be a top 10 team in SH%, if not top 5.